AND THE VISNUDHARMOTTARAPURÄNA
BY ALEXANDER LUBOTSKY
IN A RECENT ARTICLE I TRIED TO DEMONSTRATE THAT One who builds such a temple in accordance with the the prototype of the temple described in great prescriptions and always worships all gods is known äs detail in the third khanda of the Visnudharmot- Cakravartin [the sovereign of the world] in the begin-tarapuräna äs the "Sarvatobhadra" temple is the ninS of the Tretäyuga; he resides in heaven äs long äs famous Gupta temple at Deogarh (in Jhänsi Bis- he wishes' and then he attains con™on with Visnu. r τ τ Ά \ ι ΤΊ · . , , . . The man who has once worshipped all gods in this tem-trict of U.P.).1 This temple, also known m the . . , „ , . , c rr · ·η · ι ttu ur ple with all kinds of offenngs will certamly get the ben-scholarlyliterature äs the Dasävatära temple ( the efitofthegiftofthethreeworldsandwithoutanydoubt temple of the ten avatäras of Visnu"), was com- win attain whatever he desires. As soon äs the very beau-pleted in the first half of the sixth Century A.D.2 tiful temple with its Cakras and Patäkas becomes visi-and is a masterpiece of the art visi-and architecture ble, calamities disappear. Undoubtedly, anyone who of the Gupta age. Its magnificent reliefs and the enters it is not susceptible to disease, sudden death, lavishly adorned doorway have been studied time calamity. Demons have no power over him. . . . The and again, and photographs of it may be found king in whose dominion this temple is built remains in in every handbook on Gupta art. the heaven of Indra and rejoices for a long time
In spite of the fame of this temple, its icono- No harm should be done to this divine obJect·The kinS graphic program is still unknown, and several fig- or his representative who does this goes to the terrible i · i i i· r · - i -r i hell together with his sons, cattle, and relatives, and ures depicted on the reliets remain unidentmed. , . r. . . ,. ... . , . . 4
r his position m this world will certamly vamsh. In the present article I intend to show that the
description of the Deogarh temple found in the The temple is so precisely described and its Visnudharmottarapuräna (henceforth, the ab- merits so eloquently praised äs to create the im-breviation VDh will be used for the third khanda pression that the author had seen the Sarvato-of this text) may help to elucidate some obscure bhadra temple himself. The fact that he did not points in the iconography of this temple. include this temple in the classification of a hun-dred temples but described it in a separate chap-ter also points to its special importance for the Form ofthe Sarvatobhadra and Deogarh Tempks author of the VDh. He defmitely wanted to
indi-cate that the Sarvatobhadra temple is something The VDh, composed in approximately the sev- unique. "This abode of Visnu . . . must be seen enth Century A.D.,3 consists of short treatises on because the man who has seen it is released from Sanskrit and Prakrit grammar, metrics, poetics, all sins and attains merit" (87.63).
dancing, singing, music, painting, iconography, The form and proportions of the Sarvato-and architecture. These treatises take the form bhadra temple are discussed in my previous arti-of a dialogue between the king Vajra, who puts cle.5 Here I shall only mention those characteris-the questions, and characteris-the mythical sage Märkandeya, tics ofcharacteris-the temple and its shrines that are relevant who gives the answers and prescriptions. The ico- to its iconography.
thus, the porticoes have three doors. At the fourth the east. If, in reality, Väsudeva faces the west, door, behind the portico, is a shrine (probably a the whole scheme should be turned 180 degrees. niche is meant). Finally, at every corner of the The Deogarh temple is oriented to the west, main temple, at the junction of the porticoes, slightly turned toward the south in such a way stand temples with two shrines each. that the last rays of the setting sun could shine The description of the Sarvatobhadra temple on the image in the main shrine. The orienta-suits the remains of the Visnu temple at Deogarh tion to the west was not unusual for temples with
(see fig. l for its ground plan) if we take into a "dark" shrine.6 The so-called Varäha temple at account that the superstructure and the platform Deogarh, which is similar in form to our temple of the Deogarh temple have been ruined, so that and contains niches with almost identical reliefs,7 the small temples on the platform and those at is facing east. The direction of the niches is also the junction of the porticoes are no longer ex- the opposite, but their order during circumam-tant. Only a few pillars of the porticoes are left, bulation by the devotees is the same. Going clock-and these are kept in a godown near the temple. wise from the entrance of the shrine, we first see
a relief with Visnu liberating the King of the Ele-phants (Gajendramoksa), then a relief with Visnu Orientation ofthe Temples in the form of the two sages Nara and Näräyana, and fmally a relief with Visnu sleeping on the TheVDh (87.17-18) allows two alternatives for serpentAnanta (Anantasayana). The Varäha tem-the orientation of tem-the Sarvatobhadra temple: ple is turned 180 degrees in comparison with tem-the
Visnu temple, which is exactly what the VDh pre-The image of Väsudeva should face the east or the west scribes: "One should consider the direction of and no other direction. One should consider the Väsudeva äs the east, and the order ofthe other direction of Väsudeva äs the east, and the order [ofthe deities should be in conformity with this direc-deities] should be in conformity with this direction. don » There are different OpiniOnS concerning the date of the Varäha temple,8 but even if it was This seems to mean that the placement of the alatecopyof the Visnu temple ratherthan a pro-deities is given in the text äs if Väsudeva faces totype, my argument would not be affected.
t atha vaksyami sarvatobhadrasamjnitam caturätmä hariryatra kartavyo jagatlpate
I shall now teil you about the temple known äs Sarvato-bhadra, where Hari in his fourfold aspect should be placed, o king.
The term caturätmä is used here in the sense of caturvyüha (of four aspects, emanations), which refers to the quadripartite divine nature of Visnu-Krsna, the key notion ofthe so-called Päncarätra school of thought. The principal cult figures of this school are four deified heroes, the kinsmen Väsudeva Krsna, his eider half-brother Samkar-sana, Pradyumna the son of Väsudeva, and Aniruddha the son of Pradyumna.
FIG. 1. Ground plan ofthe Visnu temple at Deogarh. Drawing by Mrs. The VDh (chapters 44, 47, 54, 85) elaborates T. Wezel-Ignatova. Reconstructed portions indicated by dotted lines. the vyüha doctrine into a coherent iconographic
Iconographic Program ofthe Sarvatobhadra Temple
System of Symbols. Each of the four vyühas of ciated with sakti (creative power), holds a leath-Visnu has a particular appearance and banner er shield and a sword, and faces the north; his
(dhvaja) and is associated with a quality (guna), banner is an antelope.
attributes, and direction.9 All these qualities and According to Päncarätra doctrine, Visnu as-attributes relate to the function of the vyühas in sumes different aspects in order to save, destroy, the Universe. The following features are relevant preserve, and create the world. At the same time for the iconography of their images: all these vyühas are one and the same god, Visnu.
1. Väsudeva, the most important of the four, In sculpture this idea of the unity of the four has a "gentle" face. He is associated with bala vyühas is reflected in caturuyüha images of the (force), holds a mace and a discus, and faces the Kusäna period,10 where Visnu is represented äs a east; his banner is Tärksya (Garuda). multiple deity consisting of four figures. In the 2. Samkarsana, the emanation of the destruc- Gupta period the iconographic development of tive force, has the face of Narasimha (man-lion). the Päncarätra doctrine resulted in the image of He is associated with jnäna (knowledge), holds a Visnu Visvarüpa, an image with many heads and pestle and a plough, and faces the south; his ban- figures representing the whole Universe. An icon-ner is the Täla tree (a fan palm). ographic description of Visnu Visvarüpa is given 3. Pradyumna, the emanation of the preserv- in VDh 83, and one unfmished Visvarüpa image ing force, has the face of Kapila (a sage who was found in Deogarh.11
preached the Sämkhya philosophy, äs stated in The VDh description of the Sarvatobhadra VDh 78/1.5). He is associated with aisvarya (sov- temple attempts to present the unity of the four ereignty, supremacy), holds a bow and an arrow, vyühas, that is, the unity of the Universe, not in and faces the west; his banner is a makara (a myth- one image but in one temple, its four sides being ical aquatic animal). dedicated to the four vyühas and the other shrines
4. Aniruddha, the emanation of the creative to their consorts and attendants, their attributes force, has the face of a varäha (boar). He is asso- and banners (see fig. 2).
TRIVIKRAMA Välrritk-i Vyäsa Päoupata (+Moon) VSyij W E S T (+Saturn) Varuna ASVASIRSA |··· n ^ L) [] ubha Π Vanara515 T _, , r . Indra (+Sun) SaraD t '2 ^ Vedas yoga Särnga Γ Q '
Sähkhya Trisülaln" ί Se Π Padma Aparäjitä NARASIMHA
PLACEMENT OF THE DEITIES In tne corner temples situated in the northeast and so on, one should put Asvasirsa [with the head of a horse, The placement of the deities is given in concen- northeast], Varäha [boar, southeast], Narasimha [man-tric circles, beginning with the main shrine and iion> S0uthwest], and Trivikrama [making three steps, then moving further to the periphery. The or- northwest]. (87.24)
der of the deities is always pradaksina, or
clock-wise from the east: Samkarsana has the face of Narasimha (see "Iconographic Program of the Sarvatobhadra Now listen to the successive placement of the gods in Temple" above), which explains the position of the temple, o king. The image of Väsudeva should face the Narasimha shrine in the southwest (cf. also the east or the west and no other directions. One should 78/2.7: "Hari in his Samkarsana form assumes consider the direction of Väsudeva äs the east. The or- the form of Narasimha")
der [of the deities] should conform to this direction. Abou[ Agvagir?a (= Asvasiras, Asvagriva, Lord Samkarsana should beplaced with his face to the R Hayagriva) it is said (80.4-5) that he south. Then one should place Pradyumna with his face , ' , , , ' ° ' . , _ '. . , . ., .. J A · j.ji, vu u· f ,. ,-u *u should be represented with rour 01 his eight to the west and Amruddha with his face to the north. r 6 (87 17-20) hands on the heads of the four personified
Ve-das because long ago the VeVe-das had been re-turned by this god from Rasätala, where they were _ _, brought by two demon chiefs. His position in the PLACEMENT OF THE CONSORTS , , , · . ,- , ,,
east thus corresponds to the position 01 the Ve-The eight consorts of the four vyühas are placed das at the eastern side on the platform (see "Pan-at the junction of the porticoes. The scheme is eis around the Pl"Pan-atform" below). On the other given in the pradaksina order, beginning from band, his northern orientation contradicts 80.3: Väsudeva. Note that right and left in the follow- "Know that the god having the head of a horse is ing description are indicated from the point of a part of Samkarsana"; one would rather expect view of the image, not the viewer. the Asvasirsa shrine at the southeast.
Likewise, the southern orientation of the As to the shrines at the junction of the porticoes, about Varäha shrine is surprising because Aniruddha which I told you, one should put Laksmi to the right of (= north) has a Varaha face> and we learn from the eastern portico To the left of the southern portico 79U ^ Varaha is identifled with Aniruddha. one should place Nidra, and Kälaratri to its right. To „ ι ι 1 r 4 - i
, , r r £ . , ., . „., One would therefore expect Asvasirsa to stand the left of the western portico one should place Sid- . , r , , , ' - , . , dhi, and Rati to its right. One should put Kirti to the m the southeastern corner and Varäha m the left of the northern portico, Sarasvati to its right, and northeastern corner. Possibly the text is corrupt, Pusti to the left of the eastern portico. (87.21-23) and the order of Asvasirsa and Varäha should be
changed. Alternatively, the placement of the gods Consequently, according to the prescription of in *e Sarvatobhadra temple may not conform
u irr-vu .u r- -r r -, j to the later System of identmcations. the VDh, the rirst wife 01 every vyuha Stands to '
his right, the second wife to his left. The position
of some of the goddesses corresponds to their ·ρΗΕ "GIRDLE" description in other parts of the VDh. For
in-stance, Rati is the first wife of Kämadeva (73.21), The VDh also indicates the iconography of the who is identified with Pradyumna (52.13); Yama "g'rdle" (the cloister of twenty-four temples on and his consort Dhümornä are identified with the Platform):
Samkarsana and Kälaratri (51.7,12). Also the
position of Nidrä conforms to her description in The ^irdle should consist of little connectin§ temPles' i,0 00 _ , ι . ι ι· ι ι τ ι · i · i dedicated to twenty-four divmities [attnbutes and com-73.23-24, which dictates that she be depicted with . „,. , ~- , r^ j τ ό -, i_ r i nr panions of Visnu]: Tarksya [Garuda], Sankha [conch], her left band on her hip, Samkarsana on her nght Padma [lotusj;Cakra [discus], Längala [plough], Mus-and Väsudeva on her left. ala [pestle] f ^^^ [serpent], Gadä [mace], Täla [fan
palm], Trisüla [trident], Särfiga [bow], Sara [arrow], Makara [aquatic animal], Parasu [axe], Mudgara [ham-CoRNER TEMPLES mer·'' Carman [leather shield], Päsa [noose] ,13 Pattisa
As with the eight consorts of the four vyühas, Now listen to the [placement of the] deities on the band this list seems to begin to the right of Väsudeva. around the platform in due order: Gäyatri, the four We then arrive at a scheine that would match the Vedas' Vaisnavi Aparäjitä, Mrtyu, Käla, Yama, Danda, iconographic program (flg. 2). The three animals Kavaca' Sara' Sänkhya, Yoga, the Päncarätra and mentioned in the list stand directly to the right Pä*uPata doctrines· vYasf'Valmiki> Märkanda and the _ . _, „_ , ,_, ι \ ι ι · ι <-?!·- mahabhütos (five great elements), one after the other. of the vyühas: Tarksya (Garuda), the bird of Va- f«7 37-W
sudeva to his right, Ananta to the right of
Samkarsana, and Makara, the banner of Pradyum- Ag ·η ^ casg of ^ comer t les> this list na, to his right. The banner of Samkarsana (the of ^entyfour deities on the band Starts from the Täla tree) is also situated at his side (south). northeast, enumerating first the six eastern pan-As to the attributes of the vyühas, the plough els> followed by the six southern panels, and so and the pestle, the attributes of Samkarsana, are on In this way> we haye in the east four Vedas> placed to his left. Likewise, the bow and the ar- ^ο to the left of the staircase and ^ο to the right) row, the attributes of Pradyumna, are placed di- plus Gayatri; the personified sacred verse Rgve-rectlyto his left. The attributes of Aniruddha, the da3.62.10, "the motherof the Vedas" (VDh60.4), sword and the shield, are situated at his side and Vaisnavi Aparäjitä. Gäyatri is associated with (north) but not directly to his left. The position Sarasvatl, the goddess of learning, whose face, of the attributes of Väsudeva is unexpected. At according to VDh 64.3, 5, represents the Sävitri his left are his jewel and chaplet but not the dis- mantra (= Gayatri) and whose four hands repre. cus and the mace, which are situated to the south, sent the Vedas Aparäjitä is described in the VDh the side of Samkarsana. as a diyine mother! one of the wives of the great
god Tumburu (chapter 66). She definitely be-THE "TusK" SHRINES longs to the realm of death, the southern direc-, „ direc-, tion. Consequentlydirec-, the most northern of the The placement of the deities m the tusk shrmes eastem lg .g a Hnk ^ ^ north> whi at both sides of the staircases is determmed by most southem one is a link with the south. the direction of the staircase. The guardians, the Jn ^ south wg first sge three ds of so-called Dikpäla(ka)s, of the east are situated at and destruction_MrtyU; j^ and Yarna_and> the eastern staircase, and so on: ,, ., . , ' c ^, . · . , ·
on the other side or the staircase, their at-tributes—Danda (staff), Kavaca (armor), and Listen now to the order of the [gods in the] tusk shrines, gara (arrow) ^ scheme § full wkh ^ o kmg. Indra and Agni should be put at the east, at the , . . , „ „ ,
,°-, , .,. . . r .. , . .. destructive character or Samkarsana. south Yama and Nirrti, at the west Varuna and Anila , ' · · , ' , · · ,
[Väyu], and at the north Dhanesvara [Kubera] and In the west are the phüosophical Systems Mahesvara [Siva]. Eight planets should be put at the Sänkhya and Yoga, then the Pancaratra and the tusk shrines: Äditya [sun] and Sukra [Venus] in the Päsupata Systems at both sides of the staircase, east, Kuja [Mars] and Asura [= Rähu, who causes the then two sages: Vyäsa, the mythical author of the eclipses] in the south, Sanaiscara [Saturn] and Sasäflka Mahäbhärata and the Puränas, and Välmiki, the
[moon] in the west, Candraputra [Mercury] and author of the Rämäyana. The western side thus Brhaspati [Jupiter] in the north. At the "tusk" shrines underscores the function of Pradyumna as an of Väsudeva [that is, at the eastern staircase] should be ascetic and protector of the Universe (VDh 78/ placed two doorkeepers of Väsudeva, Subhadra and Va- -^ g\
subhadra, the lords of the three worlds; then Äsädha ' Jn Λ{, nor[h [& situated the Märkanda and Ya natata, [the doorkeepers] of Samkarsana, , ,,- , , . . , . . . _ . ΛΤ^.\
, ,,, , , IXT·· r i i i ' ' n' r (= Markandeya), the mythical author oi the VDh, should be made; Java and Niiaya, [the doorkeepers] of , , '',:.,_ ,,.' . . ... Pradyumna; Ämoda and Pramoda, [the doorkeepers] and the ^häbhütas (five great elements), which of Aniruddha. The placement of the gods and the door- are clearlY associated with Siva, bemg his five keepers in the tusk temples that I have described to faces (VDh 48.7-8), and consequently with the you is known to be of many merits. This is what one north, the direction of Siva.
should know about the guardians of the quarters of
the world and about the planets. (87.29-37) The iconographic program of the Sarvato-bhadra temple can be summarized as follows.
Ev-„ π ery side of the temple is dedicated to one of the
side, the first wife to their right and the second VDh 87.17: the direction of Väsudeva should be to their left. Their attributes are situated in six considered the east. Doing so yields the follow-small temples on every side of the platform. The ing combinations, which conform to the prescrip-panels around the platform are adorned with tions: Gajendramoksa-Samkarsana-south, different deities, sages, and deified abstract no- Naranäräyana-Pradyumna-west, Anantasayana— tions, which are associated either with vyühas or Aniruddha-north. The Varäha temple presents with their cardinal point. The corner shrines are an actual example of such an orientation. occupied by four manifestations of Visnu. In gen- The following discussion of the reliefs aims at eral, this placement scheme conforms to the sys- finding additional evidence for their dedication tem of identifications found in the Pratimälak- and at identifying some of the figures. In com-sana, the part of the VDh that describes image- parison with previous attempts, we now have two making. advantages. First, since we assume that the au-thor of the VDh had seen and admired the Deo-garh temple, his prescriptions were probably in-TheDeogarh Temple fluenced by the iconography of the Deogarh
reliefs. The use of iconographic descriptions from We may now turn to the iconography of the the VDh to identify the obscure figures on the Deogarh temple, of which only the doorway and Deogarh reliefs thus becomes more legitimate three reliefs are still intact. The panels around the and more valuable than before. Second, we shall platform have become ruins. Only several frag- compare the reliefs of the Varäha temple with ments with scenes from the lifeofKrsna and Räma those of Deogarh, which, to my knowledge, has are preserved, but these are not mentioned äs such hardly ever been done. The evidence of the by the VDh. The original place of these fragments Varäha reliefs is especially important when a part is generally unknown, so that it seems useless to of the Deogarh relief is damaged. It is irrelevant speculate on their Interpretation within the ritual which of the temples is the prototype: in both scheme of the VDh. The only means at our dis- cases the Varäha reliefs provide invaluable infor-posal to determine the iconographic program of mation.
the Deogarh temple is analysis of the reliefs.
Before we proceed with a detailed description Τηΐ, LALÄTABIMBA
of the reliefs of the Deogarh temple, letme state .., ·, , i- r-/r· o\. , . Γ. r We start with the relief (fig. 3) above the door- ι ι ι at the outset that the iconographic proe:ram or . , .. . T ._ . r . . . .
. . . · Γ i i r way (the so-called Lalatabimba), which depicts this temple seems to agree with the scheme or - , . ' . . . ' , . . T . .
, r, , , , i · r · ι ι · Visnu sitting on the serpent Ananta with Laksmi the Sarvatobhadra temple, its iour sides beme ·' . , .°_ _, r .., . r.,.
... , . r _, r,T. „, ,. ° caressmer ms foot. Two manifestations of Visnu— dedicated to the iour vyühas of Visnu. The rehef ,T . ° , .. . . T7._ ., "'
. . . 1 - 1 1 · ' \r· · · Narasimha (man-hon) a n d Vamana (dwarr)— above the doorway, which depicts Visnu sitting ' . . . . , . „ . , '
. . ' . . T , _ ' ' . . . ° appear to his nsrht and left, respectively. I he on the serpent Ananta with Laksmi and his two .rr. ... ° ._ . r . .'
.,- · χτ · ι ι τ τ- · i choice of the mamiestations seems sigmncant. mamfestations, N arasimha and Vamana, is prob- T . , „ . _, . . . ° .
, . . ,. , ... j „. . ,· r In the scheme of the Sarvatobhadra temple ably dedicated to Väsudeva. The northern relief, XT . . .„... ,. . . . r
' ... . . . Narasimha and Tnvikrama10 are placed in two representing Visnu äs a warnor hberatina: the '. . . , , r . . . . τ..Γ r ι τ-,ι ' Ί //^ · ι τ \ corner shrines behind the temple and are invisi-King oi the Llephants (Gaiendramoksa), con- . . . . , . ,
.,° . , . J r r. Ί ble to the devotee approaching the mam en-forms to the destructive power of Samkarsana. XT . , , ..° . . . T . . .,. · , r r ' XT ' trance. Now with these two manifestations depict-In the east is Visnu in the rorm of two sages, Nara . ... . . _ , _ , . . „ !,
1XT- _ '', . 1 1 · ed next to Väsudeva o n t h e Lalatabimba, a l l four and Narayana, whose ascetic power clearly points ... . _ . . '
„ , . . ,, . , mamiestations are facing the devotee.
to Pradyumna, the preservmg aspect of the god. _,. , ,, , ° . . _ „. ,, , ,· r , i · i r i The closest parallel to this representation of Fmally, the rehef at the southern side oi the tem- T7_ , . „ r . . ΛΓΓ^. n_ xrT rr. ,., ,
. ' .,. i · , r · Väsudeva is found in VDh 85.49-50, which
de-ple with Visnu engaged in the act of creation, .. ..„ . r,__ . i . ,'' * / A * \ scnbes dnierent Images oi Väsudeva: lymg on the serpent Ananta (Anantasayana), °
must be dedicated to Aniruddha, the creative _ , ,., . , . ,, . . .
Or he [Väsudeva] should be represented charmmg, " · · ' seated on the coils of Sesa [= Ananta]. Surrounded by
On the Deogarh rehef, Cakra and Gadä m hu- durua-gmss, and he wears the skm of a black ante-man form are missmg Instead, Väsudeva holds a lope " The antelope skm is missmg on the rehef conch m his lower left hand and a discus m his
lower nght hand
Incidentally, the way Vämana is depicted on the THE GAJENDRAMOKSA RELIEF
Lalatabimba corresponds to his description m Movmg around the temple m the pradaksina di-VDh 85 54-55 "Vämana should be made with rection, we come to the northern niche In con-short hmbs and a fat body, carrymg a staff and tradistmction to the other two niches of the Deo-engaged in study His color is the green of the garh temple, the scene depicted here is not
described in the VDh. The relief (fig. 4) shows attribute in his other left band is broken off. From Visnu äs a warrior fighting with two serpents, the parallel relief on the Varäha temple (fig. 5), which have bound the helpless elephant with where Visnu holds a conch in this band, we may their coiled bodies. Visnu sits on Garuda, hold- conclude that here too he held a conch. The two ing a mace in his lower right band and a discus serpents fold their hands m anjahmudrä, admit-in his upper right band, ready to throw it. One tadmit-ing their defeat and salutadmit-ing the victor. The chief of his left hands rests on his knee, while the serpent is hit by the discus of Visnu, which is vis-ible in the middle of his breast. The elephant offers a lotus flower to Visnu in gratitude. Note thaton the relief of the Varäha temple there are three serpents, not two.
Above the head of Visnu, on a separate slab, two couples hold a crown. These deities do not carry any attributes and cannot be positively iden-tified.
"In a semi-circular panel below the capital on each pilaster of this niche is a seated male figure holding [a] flower with a female companion on either side."16 Similar scenes are found on one of the pillars that presumably stood in front of this niche. The man and two women depicted on the reliefs of the pillar are drinking wine and are intoxicated.17 The function and meaning of these scenes have never been explained. It is well known, however, that wine and intoxication are typical for Samkarsana. For instance, VDh 85.73 prescnbes that "Bala [= Samkarsana] should be represented with eyes dilated through intoxica-tion." It is therefore reasonable to assume that the scenes on the pilasters and the pillar are meant to create the atmosphere of Samkarsana, to whom this side of the temple is dedicated.
FIG 4 The Gajendramoksa relief of the Deogarh temple.
Courtesy Kern Institute, Leiden, Coll. Vogel 30/24 TH£ NARANÄRAYANA RELIEF
According to the scheme of the Sarvatobhadra temple, the western side is dedicated to Pradyum-na, the preserving aspect of Visnu. VDh 78/1 (verses l and 5) states:
Know that Pradyumna is vairägya [indifferente to world-ly hfe, asceticism], when he assumes the body of the sage Kapila. He possesses great might through the vairägya, staying m meditation to the highest grade Thus meditatmg, the ancient preacher of the Sämkhya phüosophy is protectmg the world.
This explains why the relief (fig. 6) depicting the meditation of Visnu in the form of two sages, Nara and Näräyana, appears at the Pradyumna side of the temple.
Shastri has already pointed out that the de-FIG 5 The Gajendramoksa relief of the Varäha temple. scription of Nara and Näräyana given m chapter
of Deogarh 18 The only mmor discrepancy is that, on the back and tied m front "21A fold of his gar-accordmg to the VDh, Nara and Näräyana should ment flutters below On his nght his wife flies in sit on a chariot, which is not found on the relief the air holdmg her fluttermg garment
In the center of a separate relief on the top, This god must be Väyu, who is descnbed in four-faced Brahma sits on a lotus cushion The the VDh äs follows
way in which Brahma has been depicted here is
very similar to the image on the Anantasayana Väyu [wind] has the colorof the sky, and the sky should relief Only his clothes are different On the have the appearance of the wind Väyu, made visible, Naranäräyana relief Brahma wears "a wide sash has two arms> and hls garment is puffed up with the with plam borders and mcised cross-hatchmg wmd His Wlfe Slvä should be Placed at his left' wantmg t, i . ,.,.·, , .. c to move (eamaneccha) Pavana [= Väyu] should be made which seems to represent the yoga-patta of a yog- ,, vs , ' , , , , , ™
, . ι ,- ι ι ι ι j holdmg the ends of his garment with both hands The in. that is a band of cloth or other matenal used , , ·,? 0 f c i. » TM.same apphes to the goddess Siva of infinite beauty The m yogic exercises ,19 on the Anantasayana relief face of [he god shouH fee tumed aside and his hair Brahma wears an antelope hide As Maxwell sur- disheveled As Vayu assumes the smell, color, and so mises, the reason for this Variation may be the on Of everythmg he touches, he has the color of collyr-different symbohc roles that Brahma plays m ium because he has touched the sky The sky is also his these two rehefs On the Anantasayana relief, garment Siva is the motion, Anila [= Vayu] is
unob-structed (581-5) his wearing of the black antelope skm may have been
intended to emphasize his omniscient rsz-character, The fact that Siva is depicted to Väyu's nght and smce he is m the process of creatmg the world, where- not to hls left as prescribed by the VDh is msig-as, m the other panel, his wearing a piece of the yogm's nlflcant Her position is no doubt dictated by the Standard equipment is most suitable m a scene depict- COmposition of the relief
mg aplace of austenty, namely the asramaof Nara and „. . ,_ „ . . , , ,, XTÖ 20 The male ngure of the other couple holds an
Näräyana20 . , , , , , , , ,
object in ms nght hand that has not yet been iden-tified It seems to be a conch, its openmg turned Brahma is flanked by celestial couples, which to the viewer The conch of Visnu on the have not been identified To the nght of Brah- Gajendramoksa relief of the Varaha temple (fig mä is a male figure holdmg his garment, "worn 5) has been depicted m the same fashion The
left hand of the deity rests on his thigh His con-sort sits at his left, her nght hand on his back and in her left an upside-down lotus
This celestial couple probably represents Varuna and Gaun, who are described in the VDh as follows
Varuna, the lord of the sea animals, must be depicted on the chariot yoked with seven geese He wears white garments and resembles a glossy lapis lazuli [m color] He has four arms, a somewhat hanging belly, and is adorned with a pearl necklace and all Ornaments At his left side one must make a banner with a makara on it and a white umbrella over his head His wife Gaun, beautiful m all hmbs, has two arms and sits on his left lap In her left hand she holds a blue lotus, her nght hand is on the back of the god In the nght hands of the god there should be a lotus and a noose, and in the left hands a conch and a ratnapatra [vessel of jewels] (52 1-5)
There are two major discrepancies between the description of the VDh and the relief there is no chariot yoked with seven geese, and Varuna FIG 6 The Naranäräyana relief of the Deogarh temple has two mstead of four arms Both discrepancies
prescribed by the VDh for many gods, but they in VDh 85.37, and Varuna is associated with do not appear on the Deogarh reliefs. The VDh Pradyumna in VDh 52.13.
also prescribes four arms and four attributes for On the left pilaster of the Naranäräyana niche all major gods, whereas in Deogarh these gods we find Gaja-Laksmi (Laksmi with two elephants) are generally represented with two arms and only depicted in füll agreement with a description in one attribute. Except for a relief of Ganesa on VDh 82.3-8. Possibly she is positioned on the the left pilaster of the Anantasayana niche, only Pradyumna side of the temple because in this Visnu appears with four arms. The practice of representation with two elephants she is "the assigning to adeity two arms insteadof four con- mother of the entire world" (82.1). The right forms to a prescription found several times in the pilaster is rather worn, and it is therefore diffi-VDh: a deity who normally has four arms should cult to identify the figures. It is unlikely, howev-be depicted with only two arms when appearing er, that they represent "an amorous couple," äs near Visnu (e.g., 82.2 about Laksmi and 54.5 suggestedby Vats,22since such ascene wouldde-about Garuda). Moreover, in the iconographic stroy the serene atmosphere of the niche. descriptions of the VDh, a deity holds his most As all scenes in the medallions of pillar no. 2 essential attribute in his left hand, usually in his take place in a grove or under a tree,23 thus point-lower left hand, whereas on the Deogarh reliefs ing to the realm of ascetics, this pillar may have this attribute invariably appears in the deity's stood in front of the Naranäräyana niche. right hand. We observe this practice in the case
of Varuna äs well. According to the VDh, he
should hold his most characteristic attribute, the THE ANANTASAYANA RELIEF
conch, in his (lower) left hand, whereas on the On this relief (flg. 7) Visnu lies at ease on the relief he carries it in his right hand. coils of the serpent Ananta (Sesa), who
repre-The presence of Väyu and Varuna on this re- sents the primordial waters. This is the moment lief is not coincidental. These two gods are Dik- of the creation of the world, and äs Aniruddha is pälas, the guardians of the northwest and the the creative aspect of Visnu, we may assume that west, respectively, and it is in this function that this relief is dedicated to him.
they seem to appear here on the Pradyumna side, There are many correspondences between the which is the western side within the orientation Anantasayana relief and the description of the scheine of the temple. The doorkeepers of god Padmanäbha ("with a lotus emerging from Pradyumna are identified with Varuna and Väyu the navel") in chapter 81 of the VDh:
Sesa m the form of a serpent should be lymg amidst the waters. His head is dazzlmg because of the great jewels adornmg his hoods. The god of gods should be made sleeping, with four arms. One foot of the god should rest on the lap of Laksmi, the other should he on the coils of Sesa. One hand should be stretched out and rest on his knee, the second should be near the navel, the third should support his head, and the fourth should hold a bunch of Santäna flowers.24 On the lotus, emerging from the pool of his navel and con-tainmg the whole earth, should be Brahma, depicted äs above. Madhu and Kaitabha [two demons] should be made near the stalk of the lotus. Near the snake there should be the weapons in human form.
The Deogarh relief deviates from this descrip-tion in two details. First, the stalk of the lotus does not visibly emerge from the navel of Visnu but rises from behind his recumbent figure. The sec-ond point concerns the demons and the weap-ons in human form, the so-called äyudhapurusas. FIG. 7. The Anantasayana relief of the Deogarh temple The tWO figures Standing near the lotUS Stalk are
prescribed by the VDh, but Garuda and a wom- literature and drama of the Gupta period. Sec-an (probably Sec-an attendSec-ant of Laksrm) ,25 whereas ond, the tiara on the head of the third figure does the six figures below the serpent are four äyudha- not take the form of a snail spiral, äs Gail as-purusas fighting with two demons on the left sumes.30 The drawing by Vats (fig. 8) shows that side.26 The lady to the viewer's right is Gadädevi, the circles of the tiara are rather concentric, so the personified mace, and her neighbor is thattheheaddresshastheformofahideshield.31 Cakrapurusa, the personified discus. Their iden- Considering the form of bis tiara and the pecu-tity can be established on the basis of their pecu- liar (leather?) band around his hips,32 I would liar hairdo: Gadädevi's hair is done in the form venture to suggest that the man is Carman, the of the cogged point of a mace, while above the personified hide shield. This identification seems forehead of Cakrapurusa is part of a discus. The further to be confirmed by VDh 85.27, which practice of distinguishing the äyudhapurusas in states that "Carman should have the appearance this way is prescribed by the VDh: "One should of Cakra, [while] Khadga [= Nandaka] should somehow show the real forms [of the weapons] be tall." On the Deogarh relief, Cakra and Car-on theheadsof Cakra and the others" (85.27). It man look very similar; indeed, theywear thesame is therefore reasonable to assume that the other armlets in the shape of a coiled snake and the two äyudhapurusas are also characterized by an same necklaces (see fig. 9), while the man with emblem on their head, and not otherwise, äs was the sword is taller than the other two. Therefore, traditionally assumed (these figures were seen äs the third and fourth figures seem to be Carman Dhanus [bow] and Nandaka [sword], the latter and Nandaka. These two figures occupy the cen-because of his sword and the former cen-because he tral position on the relief, possibly cen-because the "positions his right leg äs if it were a bow") ,27 sword and the shield are the emblems of Anirud-Recently Gail has suggested that the headdress dha, to whom the niche is dedicated.
of the third figure has the shape of a conch and Let us now take a closer look at the group of that, consequently, this must be Sankha; van Kooij five gods at the top of the Anantasayana relief. identified the fourth figure äs Padma because of In the center is Brahma on a lotus, flanked on a lotus in his hair.28 the left by Siva with Pärvati sitting on his usual
These identifications are tempting but, unfor- vähana, the bull, and on the right by Indra on tunately, not without problems. First, äs Gail ob- his elephant Airävata. In his right band Indra serves,29 the lotus äs an attribute of Visnu does probably holds a vajra, which is now broken off not appear in sculpture before the seventh cen- both on our relief and on the Anantasayana re-tury A.D., so that it is better to stick to the tradi- lief of the Varäha temple. According to the VDh tional view that the fourth figure is Nandaka, es- (50.5), Indra should hold a vajra in one of his pecially äs Nandaka often occurs in Sanskrit left hands (presurnably his lower left band, äs the
FIG. 8. The hairdo of Carman. Drawing after M S. Vats,
by the VDh for a (lower) left hand appears on the relief in the right hand.
Whereas the identity of the central three fig-ures on the panel— Brahma, Siva, and Indra—is beyond any doubt, the identity of the other two deities is still under discussion. The figure to In-dra's right has generally been identified äs Kärt-tikeya (Kumära, Skanda) because of his vähana, the peacock. This view has recently been chal-lenged by van Kooij, who states (together with Debala Mitra) that the vähana of this god is not a peacock but a goose and that, consequently, this FIG. 10. Siva and Sürya on the Anantasayana relief must be Varuna.33 The bird, indeed, lacks any of the Deogarh temple. Photo: Professor K. van Kooij. characteristic of the peacockand looks verymuch
like a goose, but this is insufficient to prove that other left hand rests on the back of his consort the god is Varuna, because, äs van Kooij himself Saci). In conformity with the general practice of has shown,34 geese and peacocks are not always the Deogarh temple (see "The Naranäräyana clearly distinguished in the iconography. On our Relief above), Indra has only two arms and holds relief the attribute of the god cannot be deter-the attribute in his right hand. mined because his right hand is broken off, but
To my knowledge, the attribute in Siva's right on the Anantasayana relief of the Varäha temple hand has not been identified. It is a small object, (flg. 11) the god holds a spear (or a long sword) and Siva is depicted äs if looking at it (fig. 10). It in his right hand. This is a decis.ive argument seems probable that it is a mirror. From VDh against Varuna, äs the spear is Kärttikeya's at-55.2-3 we learn that "Siva should be with four tribute. According to VDh 71.5, Kärttikeya has hands: in his right hands one should put a rosary four arms, with a cock and a bell in his right and a trident, and in his left hands a mirror and hands, a banner and a spear in his left hands. a blue lotus." Here again, the attribute prescribed Once again the VDh puts the main attribute in
the god's left band, most probably in the lower abandoned since the figures thatwere identified left band, äs the most appropriate place for the by van Kooij äs Varuna and Väyu seem to be Kärt-banner is the upper band. tikeya and Sürya. Kärttikeya is the son of Siva and
The figure to Pärvati's left (fig. 10) has always leads the army of gods (see VDh 71.7); he prob-confused scholars, who either refrained from any ably belongs to the Rudras but is never men-identification or called him a Marut, a Vidyädhara, tioned äs their representative. Sürya is often seen a garland-bearing figure, generally without any äs the representative of theÄdityas (see VDh 72.7: elaboration. Only van Kooij has discussed bis "the twelve Ädityas should have the form of identity at some length, hesitatingly proposing Sürya"). Accordingly, the group of Vasus is ab-to identify him äs Väyu: sent from the relief.
Above the Naranäräyana relief are Väyu and The only detail which may give some indication is the Varuna in the function of Dikpälas, the guard-fluttering piece of cloth in his right hand In early ians of the northwest and the west, respectively. iconography, from the Kusäna period onwards . . . , It is therefore probable that the gods represent-Väyu's main characteristic is the inflated garment.35 ^ Qn the Anantasayana relief are Dikpälas äs well.
Siva is known äs the guardian of the northeast-Several indicators, however, render this identifi- ern direction and Indra äs the guardian of the cation improbable. First, the garland he holds in east. Kärttikeya, who is the son of Siva, also be-his right hand is hardly a piece of be-his garment. longs to the northeast, while Sürya represents the Second, the god has a halo around his head, east and is situated, together with Indra, in the which is clearly visible on the relief of the Varäha eastern tuskshrine of the Sarvatobhadra temple. temple (fig. 11). Finally, in my opinion, Väyu is As we have seen above, the Anantasayana relief represented on the Naranäräyana relief (see "The is directed to the north within the orientation Naranäräyana Relief" above), where he has a very scheine of the temple, and the presence of Siva different appearance. äs the Dikpäla of the northeast is understandable. I propose to identify this god äs Sürya, whose Note that he is the only god on the relief who is characteristic features are the raised hands with accompanied by his wife, äs are Väyu and Varuna (mostly double) reigns and a halo. The VDh on the Naranäräyana relief. It remains unclear, (67.2-4) describes the image of Sürya äs follows: however, why the Dikpälas of the east are
repre-sented on the Anantasayana relief.
The Sun should be made with a beautiful moustache The northern orientation of the niche is fur-and beof the vermilion color. He should wear the dress ther accentuated by the reliefs on its pilasters. of the Northerners, be adorned with all Ornaments, and On the pilasters to the viewer's left is Ganesa, be of good appearance. He should have four hands and another son of Siva. On the opposite pilaster sits be very lustrous. He should be covered by an armor i r· -«.u u· 4- · u· · u» u j
, ; . „ , ; , a male fisrure with an obiect in bis right hand, and wear a girdle known äs Yaviyanga. In his left and , . . . σ , , . . . ?. . _ . . , , , f , u u u!. -r i · 1 - 1 1 which is worn and cannot be positively identined right hands he should hold beautiful reigns, which have . r '
the form of garland threads going up and are adorned but it looks more like a weapon than flowers, äs by flowers. suggested by Vats. The deity looks at a smaller figure to his right, and this fact may help identify him äs Nandin, an attendant of Siva, who is de-The description of the reigns suits the relief, but scribed in VDh 73.15-17 äs follows:
Northerners' dress (who used to cover their
body)36 is absent. The halo around Sürya's head Nandin should be made with three eyes and with four is mentioned in Varahamihira's Brhat-Samhitä longarms. His color is vermilion. He wears a tigerskin. (57.48). One should make him with two hands holding a tri-What is the meaning of these gods äs a group? dent and a bhindipäla [a short javelin]38 with his third According to van Kooij, they represent the Heav- hand on his head and with his fourth hand makinS a en of the Thirty-three Gods, Varuna being threatening S^ture. He should be represented äs
, , , „ , -r , f,'. , watching people commg from a distance. present äs the leader of the Ädityas, Siva äs the
leader of the Rudras, Väyu äs the leader of the
(Ganesa), and in the vicinity of Väsudeva one teniple. He devoted a whole chapter to a thor-shouldplace SesaandTärksya (Garuda) (86.139). ough description of its iconographic program The triad Siva-Ganesa-Skanda is also mentioned and incorporated many iconographic elements in 86.35 (ganesarudraskanda-). This prescription into the System of identifications that can be accords with the reliefs of the Deogarh temple. found in the Pratimälaksana, the part of the VDh
that describes image-making.
The iconographic program of the Deogarh Conclusion temple is closely connected with the Päncarätra
Notes 11. Maxwell, Visvarupa, 233ff.
I am indebted to Prof. Karel van Kooij and Dr. Ellen 12. In the edition by Priyabala Shah, VDh 73.24cd Raven for commentmg upon an earlier version of this reads väsudevas tathä väme mmkänam ca tathä sutah. paper. I would also like to thank the Fnends of the Kern Päda d does not make sense. The readmg of the Institute and, agam, Karel van Kooij for kind permis- Venkatesvara Press edition maikänesä tathä and the sion to reproduce the photographs. readmg of ms. B maikänämsäs tathä suggest that the goddess Ekänamsä must be meant here. She is 1. Alexander Lubotsky, "The 'Sarvatobhadra' Tem- descnbed m VDh 85.72 äs standmg between Räma ple of the Visnudharmottarapuräna and the Visnu and Krsna, with her left hand on her hip and a Temple at Deogarh," m Ritual, State and History in lotus in her right hand (the position of Ekänamsä South Asia Essays in Honour ofj C Heesterman, ed. between Räma and Krsna also follows from VDh A W. van den Hoek, D. H. A. Kolff, M. S. Oort 86 26). The verse is evidently corrupt (Stella
Kram-(Leiden: Bnll, 1992), 199-221. nsch, TheVishnudharmottara[PartIII] A Treatiseon Indian Paintmg and Image-Making, 2nded. [Calcut-2. See Joanna Gottfried Williams, The Art of Gupta In- ta: Calcutta Umversity Press, 1928], 97), but the
dia Empire andProvmce (Pnnceton. Pnnceton Um- meamng must be "And so also Ekänamsä [should versity Press, 1982), 132 and n. 98, for a discussion be represented]."The corruption must have orig-of the date. inated when a scribe erroneously connected väma
ekänamsä into vämaikänamsä 3. Lubotsky, "'Sarvatobhadra' Temple," 201f
13 Note that m the edition by Priyabala Shah, carma 4. This and subsequent quotations of VDh are trans- päsam is erroneously written äs one word.
lated from Visnudharmottara-puräna, third hhanda,
ed. Priyabala Shah, 2 vols., Gaekwad's Oriental Se- 14 In spite of the readmg of all manuscnpts m VDh ries, nos. 130, 137 (Baroda: Oriental Institute, 85.30, 85 36, and here, Priyabala Shah everywhere 1958,1961). wntes Yajnatära.
5. Lubotsky, "'Sarvatobhadra' Temple " 15. Tnvikramaand Vämanaare twoformsof the same manifestation of Visnu, reflectmg the myth about 6 See Stella Kramnsch, The Hindu Temple (Calcutta: Visnu in the form of a dwarf, who asked an Asura Calcutta Umversity Press, 1946), 272. for a space equal to his three Steps. When this re-quest had been granted, the dwarf suddenly be-7. N R Banerjee, "New Light on the Gupta Temples gan to grow and, assummg gigantic proportions,
at Deogarh, "Journal ofthe Asiatic Society (Calcutta) covered the whole Universe in three steps. 5, 1/2 (1963): 37-49.
16. MS. Vats, The Gupta Temple at Deogarh, Memoirs 8. See Lubotsky, "'Sarvatobhadra' Temple," 210f. and ofthe Archaeological Survey of India, no. 70
(Del-n. 15. hi, 1952), 28.
9. J. N Banerjea, The Development of Hindu Iconogra- 17. Vats, Gupta Temple, 28 phy, 2nd ed. (Calcutta. Umversity of Calcutta,
1956), 409, wntes that the VDh assigns the north 18 B. Ch. Shastri, "Identification of a Relief Belong-to Pradyumna and the west Belong-to Amruddha, assum- mg Belong-to the Gupta-Temple of Deogarh," Acta Onen-ing that the faces of Visnu are mentioned in the taha 12 (1934)· 116-25 The myth of two brothers text in the followmg order: front = east, nght = Nara and Näräyana is also given in hhanda l, chap south, left = north, and back = west This interpre- 129, 1-19, and alluded to m hhanda 3, chap. 35, tation is repeated by W. E. Begley, Visnu 's Flammg where the origin of the art of pamüng is
associat-Wlieel The Iconography o/<AeSudarsana-Cakra (New ed with the creation of Urvasi York. New York Umversity Press, 1973), 39, n. 182
and 43, n 202, who follows Banerjea. In reality, 19. Maxwell, Visvarupa, 117. the VDh always gives the directions in the
pradaksina order, startmg from the east: east, south, 20 Maxwell, Visvarupa, 117 west, north (see "Placement ofthe Deities").
21 Vats, Gupta Temple, 35. 10. See T S Maxwell, Visvarupa (Delhi: Oxford
23. Vats, Qupta Temple, 29. Orientale dipartamento di studi Asiatici, senes mmor 12 (Naples, 1985), 683.
24. On the Deogarh relief, Visnu's hand holding the
Santäna flowers is broken off, but on the Varäha 29. See van Kooij, "Gods and Attendants," 683, n. 4. relief the flowers are still present (see flg. 11).
30. Gail, "Äyudhapurusa," 183, n. 11. 25. This figure is generally idenüfied äs Bhüdevi (cf
Vats, Gupta Temple, 15), but note that on the 31 Vats, Gupta Temple, pl. XXVII, 11. Anantasayana relief of the Varäha temple (flg. 11),
next to the head of Visnu, there is a small female 32. See Vats, Gupta Temple, pl XX, 9 figure on the coils of Sesa. She cannot but be the
goddess of the earth, so that the figure behind 33. Van Kooij, "Gods and Attendants," 686ff; Debala Laksnii is probably her attendant. Mitra, "Varäha-Cave of Udayagin—An Iconograph-ic Study," Journal of the AsiatIconograph-ic Society (Calcutta) 5, 26. See Vats, Gupta Temple, 15, n. l, andSusan L. Hun- 3/4 (1963): 100, n 2.
tington, The Art ofAnaent India (Buddhist, Hindu,
Jmn) (New York: Weatherhill, 1985), 210. 34 "Gods and Attendants," 687
27. Huntmgton, Ancient India, 210. 35. Van Kooij, "Gods and Attendants," 698f.
28. A. J. Gail, "Äyudhapurusa—Die anthropomorphen 36. See Banerjea, Hindu Iconography, 436ff. Waffen Visnus in Literatur und Kunst,"
Gedenk-schrift L. Sternbach, Indologica Taunnensia 8-9 37. Gupta Temple, 15. (1981): 183, n. 11; K R. van Kooij, "Gods and
At-tendants in the Relief of Visnu Anantasayana at 38. See G. Liebert, IconographicDictionary ofthe Indian Deogarh," in South Asian Archaeology 1983, ed. J Rehgions (Leiden· Bnll, 1976), 39.
